如何获取和比较pyspark中两个dataframe中相似列的所有值的数据类型

# 导入必要的库
from pyspark.sql import SparkSession

# 创建SparkSession
spark = SparkSession.builder.getOrCreate()

# 创建两个DataFrame
df1 = spark.createDataFrame([(1, "John", 25), (2, "Alice", 30), (3, "Bob", 35)], ["id", "name", "age"])
df2 = spark.createDataFrame([(1, "John", 25), (2, "Alice", 30), (4, "Charlie", 40)], ["id", "name", "age"])

# 获取DataFrame中的列名和数据类型
columns1 = df1.columns
dtypes1 = df1.dtypes
columns2 = df2.columns
dtypes2 = df2.dtypes

# 获取两个DataFrame中相似列的所有值的数据类型
similar_columns = list(set(columns1) & set(columns2))

result = []
for column in similar_columns:
    # 获取列的数据类型
    data_type = [dtype[1] for dtype in dtypes1 if dtype[0] == column][0]
    
    # 获取列的所有唯一值
    values = df1.select(column).union(df2.select(column)).distinct().collect()
    values = [str(row[0]) for row in values]
    
    result.append((column, data_type, values))

# 打印结果
for column, data_type, values in result:
    print(f"列名: {column}")
    print(f"数据类型: {data_type}")
    print(f"所有值: {values}")
    print()